Abstract

The integration of artificial intelligence (AI) technologies into user interface (UI) design is transforming the traditional design process. This rapid rise in AI use in user interface design has raised questions about the effectiveness of AI-driven interfaces compared to those created by human designers (Stige et al., 2023). This study aims to address these questions by conducting a comprehensive comparative analysis of AI-designed web-based interfaces and human-designed web-based interfaces across key performance metrics. Using the Technology Acceptance Model (TAM) framework, the study will examine user acceptance and adoption of AI-designed UIs based on perceived usefulness and perceived ease of use. The study will utilize a between-subjects design, randomly assigning participants to interact with either AI-designed or human-designed interfaces, while ensuring interface equivalence in functionality and purpose. User interfaces for each group will be similar in functionality and purpose to minimize confounding variables. Our target population will be general users who have basic computer literacy and experience using digital interfaces. The independent variable will be the type of interface design (AI-designed vs. human-designed). The dependent variables will be the use of usability metrics such as ease of use and task completion time), user satisfaction (using surveys and ratings of interface experience), efficiency (time taken to perform the tasks and number of steps required), and effectiveness (accuracy of task completion and achievement of user goals). The study will have participants perform predetermined tasks or scenarios using the assigned interface. Quantitative data will be collected on usability metrics, efficiency, effectiveness, and qualitative data through user feedback and satisfaction surveys. The two types of interfaces will be designed and developed to ensure that the interfaces are well-balanced in terms of design principles, features, and functionality. The study will include a pilot test of the interfaces with a small group of participants to identify and address any usability issues or design flaws. We will also validate the measurement tools such as usability questionnaires and performance metrics for reliability and validity. During data collection, we will randomly assign participants to either the AI-designed interface group or the human-designed interface group. We will have participants perform predetermined tasks or scenarios using the assigned interface type. We will then collect quantitative data on usability metrics, efficiency, effectiveness, and qualitative data through user feedback and satisfaction surveys. In the data analysis phase of the research study, we plan to use appropriate statistical analysis techniques (e.g., t-tests, ANOVA, regression analysis) to compare the performance of AI-designed and human-designed interfaces. We also plan to analyze qualitative data to gain insights into user experiences, preferences, and perceptions of each interface type. Overall, the research purpose is to contribute knowledge, insights, and evidence-based recommendations regarding the performance, usability, satisfaction, and efficiency of user interfaces designed with artificial intelligence compared to those created solely by human designers.

Download

Share

COinS